Dietary factors in gastrointestinal tract cancers – an Asian perspective

Stomach cancer is still the most frequent cancer in Eastern Asia, especially in Japan (104 000 cases in 1999 and 50 000 deaths in 2003) and Korea, although its age-adjusted rates are steadily decreasing like those observed in Western countries. In contrast, colorectal cancer is increasing and approaching the most frequent cancer in Japan (94 000 cases in 1999 and 39 000 deaths in 2003).

Within Japan, there exists an almost 3-fold difference in age-adjusted mortality rates of stomach cancer: higher rates are in the northern areas such as Akita, Yamagata and Niigata while lower rates are in the southern Kyushu, and especially in Okinawa Despite of the different secular trends, the similar geographical distributions of age-adjusted mortality are seen in colorectal cancer (r = 0.43 in men and 0.33 in women, based on 1995 mortality data of 47 prefectures).

The Japanese immigrants to the USA showed the significant decrease in stomach cancer incidence and rapid increase in colorectal cancer incidence even among the first-generation. These changes reflect the secular trends in both cancers in Japan. However, the Japanese immigrants to Brazil did not show such remarkable changes [1].

Thus, geographical and ethnic differences, trends in cancer incidence with time, and change of incidence patterns observed among immigrants all indicate that stomach and colorectal cancers are closely associated with modifiable factors such as diet.

The available evidence shows that the risk of stomach cancer is increased in individuals with Helicobacter pylori (H.p.) infection and/or smoking habit, whereas decreased in those with a high consumption of fruit and vegetables, and/or a low consumption of salt and salted food. Among these factors, our ecological and prospective studies confirmed that salt and salted food intake was closely associated with the risk of stomach cancer at both population and individual levels in Japan. An ecological study of five selected areas in Japan showed an almost linear correlation between the cumulative mortality rates up to 75 years of age (%) and the urinary salt excretion levels in 24-hour urine samples [2]. A population-based prospective study, the JPHC Study, showed that higher intake of some traditional salt-preserved food and salt per se, as well as low intake of fruit and vegetables were associated with the risk of stomach cancer [3], [4]. H.p. infection was shown to be a strong predictor for future development of stomach cancer in nested case-control study within the JPHC study [5]. In a cross-sectional study, the consumption of salted food appears to increase the risk of H.p. infection [6]. Mucosal damage induced by salt and salted food may increase the possibility of persistent infection with H.p. Dietary pattern analysis showed that the traditional pattern increased the risk of stomach cancer, while the healthy pattern decreased the risk among women only [7]. Cigarette smoking is attributable for one third of male stomach cancer in Japan [8].

The current best established diet-related risk factor for colorectal cancer is overweight and obesity, especially in men. Other risk factors are high consumption of preserved and red meat intake and low consumption of fruit and vegetables. Our prospective study showed that alcohol consumption, as well as cigarette smoking [9], and overweight [10] were associated with increased risk of male colorectal cancer. Although consumption of fruit and vegetables [11] and fish [12] showed null association, very low consumption of dietary fiber increased the risk of colorectal cancer [13]. Dietary pattern analysis showed that both traditional and Western dietary patterns were positively associated with colon (not rectal) cancer risk in women and no dietary patterns were associated with colorectal cancer risk in men [14].

Further evidence is necessary to elucidate dietary factors in gastrointestinal tract cancers, which accounted for one third of total cancer death in Japan, and to implement their prevention.